The Association of C-Reactive Protein with Subclinical Cardiovascular Disease in HIV-infected and HIV-uninfected Women

Abstract

Objective

HIV is a cardiovascular disease (CVD) risk factor. However, CVD risk is often underestimated in HIV-infected women. C-reactive protein (CRP) may improve CVD prediction in this population. We examined the association of baseline plasma CRP with subclinical CVD in women with and without HIV.

Design

Retrospective cohort study

Methods

572 HIV-infected and 211 HIV-uninfected women enrolled in the Women’s Interagency HIV Study underwent serial high-resolution B-mode carotid artery ultrasonography between 2004-2013 to assess carotid intima-media thickness (CIMT) and focal carotid artery plaques. We used multivariable linear and logistic regression models to assess the association of baseline high (≥3 mg/l) high-sensitivity (hs) CRP with baseline CIMT and focal plaques, and used multivariable linear and Poisson regression models for the associations of high hsCRP with CIMT change and focal plaque progression. We stratified our analyses by HIV status.

Results

Median (IQR) hsCRP was 2.2 mg/l(0.8-5.3) in HIV-infected, and 3.2 mg/l(0.9-7.7) in HIV-uninfected, women (p=0.005). There was no statistically significant association of hsCRP with baseline CIMT [adjusted mean difference -3.5 μm (95%CI:-19.0-12.1)] or focal plaques [aOR:1.31(0.67-2.67)], and no statistically significant association of hsCRP with CIMT change [adjusted mean difference 11.4 μm(-2.3-25.1). However, hsCRP ≥3 mg/l was positively associated with focal plaque progression in HIV-uninfected [aRR:5.97(1.46-24.43)], but not in HIV-infected [aRR:0.81(0.47-1.42)] women (p=0.042 for interaction).

Conclusion

In our cohort of women with similar CVD risk factors, higher baseline hsCRP is positively associated with carotid plaque progression in HIV-uninfected, but not HIV-infected, women, suggesting that subclinical CVD pathogenesis may be different HIV-infected women.